SYMPOSIUM: NUTRITION

An evidence based guide to weaning preterm infants

Caroline King

Difculties around weaning can be accentuated by conicting advice from health professionals, family and friends, many of whom have little experience with this group of infants. In addition mothers may experience more anxiety compared to mothers of term infants which can impact on feeding. This guideline seeks to give information based on published evidence where ever possible in an attempt to avoid conicting advice3 and to supply a recognised need.4

Abstract
There have been national guidelines for weaning infants for many decades in the UK however none have addressed preterm infants until the COMA report of 1994 which gave brief recommendations. More recent publications dont include recommendations for preterm infants and it is acknowledged that for this group there is a need for separate guidelines; however none to date have been published. In this article the evidence base around weaning preterm infants is discussed and updated guidelines are suggested. It is suggested that preterm infants should be considered for weaning between 5 and 8 months uncorrected age to ensure that sensitive periods for the acceptance of solids are not missed and to allow development of appropriate feeding skills. The exact time within this 3 month window will depend on the individual infants needs and cues. Although prematurity is usually dened as birth before 37 completed weeks, those healthy preterm infants born from 34 completed weeks onwards in general may be weaned according to the same guidelines as for term infants. To allow for a degree of motor development which may help with positioning it is suggested that it may be useful to delay weaning until after 3 months corrected age. However the assessment of infant cues are paramount when making the nal decision and each baby should be treated as an individual according to his/her own developmental readiness.

When have preterm babies been weaned in the past?

Studies carried out around 20 years ago found weaning occurred between 2 weeks and 4 months corrected age5 and by 4 months uncorrected.6 Another more recent study found two peak times for weaning 12 and 16 weeks uncorrected age, but there was a wide range.7 Preterm infants born in the mid 1990s were more likely to be on solids both at 6 and at 12 weeks corrected than term infants.8 In another study 21% of babies studied were weaned before reaching 5 kg9 as with term babies this study showed that breast fed preterm babies were weaned later than formula fed. Others have found weaning occurring around 7 weeks corrected age10; and in more recent samples 22 and 24 weeks uncorrected respectively.11,12

Guidelines for term babies

In 2001 the Scientic Advisory Committee on Nutrition (SACN), discussed the World Health Organisation (WHO) recommendation for exclusive breastfeeding for the rst 6 months with introduction of solids thereafter. It was acknowledged that this recommendation applies to populations and that the considerations for low birth weight infants would have to be met under a different recommendations.13 In 2003 the UK adopted a version of the recently updated WHO guidelines on weaning.2 WHO advised exclusive breastfeeding to 6 months, whilst the UK government adapted this to advise the delay of solid food until 6 months: while encouraging exclusive breastfeeding. It was acknowledged that this was a guideline for populations and adapted guidelines may be needed for special groups of infants. In the USA it was also felt that in certain circumstances some special groups may need to start solids at a different time from the population guidance.14

Keywords guidelines; preterm; weaning

Introduction
In this article weaning is dened as the introduction of solid food into an infants diet, also known as complementary feeding. There have been national guidelines for weaning infants for many decades in the UK however none have addressed preterm infants until the COMA report of 1994 which gave brief recommendations.1 More recent publications dont include recommendations for preterm infants and it is acknowledged that for this group there is a need for separate guidelines2; however none to date have been published. In this article the evidence base around weaning preterm infants is discussed and guidelines are presented.

Why wean
It is useful to look at the reasons why this process occurs and then evaluate how it applies to preterm infants. There are many reasons for initiation of the weaning process, not all will be covered here. Instead those factors most pertinent to the debate around weaning preterm infants will be discussed, through a comparison of term and preterm data. Iron Exclusively breast-fed term infants appear to maintain iron status until approximately 6 months,15,16 whereas those fed a cows milk formula not fortied with iron will become depleted by approximately 4 months and preterm infants by approximately 2 months.15 Maintenance of iron status therefore may be given as a reason for weaning. It is recommended that breast fed preterm

Why are specic guidelines needed?

As in-patients many of these infants receive extensive high technology care and individualised nutrition. However, this vulnerable group who are known to be at risk for poor growth receive very little general or individualised advice on weaning, particularly its timing. There is a risk that poor weaning practice could lead to nutritional compromise, which the preterm infant is poorly equipped to withstand.

Caroline King BSc SRD is a Paediatric Dietitian (Neonatal Specialist) at the Department of Nutrition & Dietetics, Hammersmith Hospital, Imperial Hospitals NHS Trust, Du Cane Road, London W12 OHS, UK.

PAEDIATRICS AND CHILD HEALTH 19:9

405

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

infants are routinely supplemented with 2e4 mg iron/kg/day post discharge, or, if not breastfed, given formula which will supply an equivalent amount.17 Post discharge formula with a higher iron level than term formula is useful for the rst few months corrected age, however after that time iron fortied formula designed for term infants appears to be sufcient to maintain iron status for the majority. There is some evidence that formula with levels as low as 0.8 mg/kg maintain normal iron status in the rst months following discharge from the neonatal unit.18 However there were small numbers of extremely low birthweight infants in this cohort who may need the higher dose in the early months post discharge to replenish their initially smaller stores. In term infants there is evidence for up regulation of iron absorption between 6e9 months allowing adaptation to lower iron intakes.19 In this study complementary foods were found not to reduce iron absorption from breast milk, the equivalent study has not been carried out in preterm infants and would useful additional information to help form appropriate weaning advice. A randomised controlled trial evaluated weaning at a mean age of 15 or 17 weeks uncorrected, with guidance on using foods which were of high nutrient density.20 The early weaning group were advised to use preterm formula to make up weaning solids, but the main milk drink was not controlled for in either group, and was either breast milk or term formula. At 6 months the early weaning group had signicantly higher haemoglobin and serum iron, although no babies in either group were anaemic. In summary; maintenance of iron intake is not a prime reason for weaning in preterm infants, as breast fed babies are recommended to receive an iron supplement until they are assessed as having sufcient iron in their diet and formula fed babies are now usually given an iron enriched post discharge formula. Energy Term infants will reach the point where milk consumption does not satisfy energy requirements and the introduction of solid food helps by providing relatively energy dense nutrition.21 They are sensitive to dietary energy density but below a certain level they cannot up regulate volume sufciently to supply needs.22 In the initial stages the weaning diet will not make a signicant contribution to energy needs but eventually, with appropriate choice of food groups, it will do. In normal development there appears to be a pattern of increase in energy density of the diet over the rst year of life.23 Reduced intake of breast milk has been shown following introduction of solids in well fully breast fed infants.24,25;though reduced growth has not been shown.25e28 There is contradictory evidence with respect to formula-fed term infants, one paper showing no reduction24 another showing a decrease in formula intake with weaning.26 Some have shown that weaning can lead to a reduction in dietary energy density29 however this need not always be the case.30 As with term infants there is some evidence that introduction of solids in preterm infants leads to a reduction in milk intake.5 In summary; for formula fed preterm infants a higher energy density feed could be given if needed, so for this group strictly speaking weaning is not needed to provide a more nutrient dense diet until later in the rst year. However for breast fed babies energy will eventually become limiting and complementary feeding will be needed to supply needs.

Oro motor development In the rst year of life a large number of anatomical, developmental and neurological changes occur which inuence feeding in term infants allowing the change from exclusive milk feeding to solids by spoon and by hand.31 Throughout the rst 2e3 years feeding skills are continuously rened, ultimately resembling mature feeding behaviour: a combination of developmental changes and experience with food aiding the acquisition of these skills.32,33 The newborn takes milk initially via suck feeding involving reexive rhythmic up and down jaw and tongue movement. Progression is made to the more mature oral patterns as the tongue is able to move independently of the jaw; a step towards aiding the oral manipulation of soft food. Then as the reexive suck diminishes lateral tongue movements develop to enable chewing of more textured food33 It is not known how much this progression is pre-programmed and how much learnt, but infants who lack the opportunity to practise skills at the appropriate time appear to be a risk of later feeding problems. Lip seal, to enable spoon clearance has been suggested as a sign of readiness to wean,1 however efcient lip seal is not usually seen until around 8 months in term infants and therefore is not a useful sign.34 The protrusion or extrusion reex of the tongue occurs in full term neonates in response to touching the tongue tip; it diminishes after 4e6 months aiding the acceptance of solids. It has been described as the tongue thrust reex1 and it has been suggested that weaning should occur once this reex is seen to diminish.1 The term protrusion reex is a better description of this normal developmental stage as tongue thrust is a commonly used term suggestive of neurological impairment. Some tongue protrusion often persists even after weaning has started, gradually disappearing with the progression of solids: in term infants this tongue movement may persist until 9 months.34 The rst solids are often approached with in-out tongue movements so some food will be pushed out of the mouth.31 In term infants lip seal develops and tongue protrusion subsides over variable periods of time, probably aided by the introduction of semi-solids by spoon. In the absence of rm evidence it seems reasonable to assume that preterm babies also learn to rene these skills once semi-solids have been introduced, rather than waiting for them to develop independently. For this reason it seems appropriate to wait for lip seal and tongue protrusion to diminish but not disappear before weaning is started. In summary it is not necessary to wait for lip seal to develop and tongue protrusion to diminish before weaning e in fact they may only mature with the aid of weaning. Development of taste The term taste is used in this section, although avour would be a better description as avour is a combination of taste and odour. Neonates are born with a preference for sweet tastes, and neutrality or aversion to others,35 however altered facial expressions are demonstrated in both preterm and term infants when presented with acid, sugar and bitter tastes.36 It has been suggested that rejection of bitter taste doesnt occur until beyond the neonatal period,37 and indeed infants up to 4 months old have been shown to easily become fully

PAEDIATRICS AND CHILD HEALTH 19:9

406

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

acclimatised to bitter tasting protein hydrolysate formula, beyond which acceptance of such formulas greatly diminishes.38 By 4e6 months a preference for salt taste has been demonstrated in term infants by some,39,40 which may coincide with the ability of the kidney to handle a higher solute load41; however others have not made the same nding.42,43 An increase in acceptance of salt may be more dictated by maturation than experience.40,44 One study which found no preference for salted food did not control for mothers possibly overriding their infants satiety cues.42 As the sucking reex diminishes the older infant begins to exert more control over feeding, willingly ingesting or refusing food. Refusal could be the result of many factors including satiety, and unfamiliar textures or tastes. From 4e5 months a term infant can express a wide variety of communicative behaviours especially at meal times e.g. vocalising eagerness when items associated with feeding are seen and communicating hunger, discomfort, food preferences, satiety and the preferred speed of delivery of food.40,45,46 These cues can be subtle so it is key that the person feeding watches the baby carefully to avoid overriding them. From around a year a child may more consciously begin to take control by refusing food, reecting their increasing autonomy.33,47,48 This is different from refusal of an unknown food (neophobia) which can develop around 12 months but is more usual at later ages. A larger variety of avours given in the initial stages of weaning enhances acceptance of a novel food later.49,50 In one study feeding of a strong tasting protein hydrolysate formula from the rst week of life increased acceptance of an alternative unpalatable novel formula at 7 months.38 Once weaning commences infants given repeat exposure to a food will increase acceptance in most cases.43 In this study infants were given the test food once a day for 10 days. Sight of food alone is not sufcient to increase acceptance. In young children it has been shown that they must also get the opportunity to taste and touch food.51,52 There is evidence that repeat exposure to an initially refused new food will increase acceptance. In toddlers it is thought to take at least 10 exposures to enhance acceptance of a new food e with a suggested frequency of 2e3 exposures per week.51,53 Facial expressions assumed to be distaste could lead to reduced exposure of a particular food,54 however this may not mean that the food will not be accepted eventually with continued exposure. Development of food acceptance is enhanced with the inclusion of the child in family meals giving plenty of opportunity to watch, learn and imitate feeding.55e57 Strong family similarities have been noted in preferred foods.58 Persistent refusal of a particular food may be due to genetic variation in taste perception which has been shown for bitter59 and sweet.60 In these situations if only a few foods are refused it will not compromise nutritional input. In one survey 19% of a sample of 95 mothers thought a wide variety of tastes was important but another 14% believed that weaning foods should be bland.61 Where carers have perceived that babies only like bland foods, or if they are not adventurous themselves they will need encouragement to be bold with the variety of tastes they give their babies. Good reviews around this area include.62e64 The fetus is exposed to amniotic uid avoured by mothers diet while in utero.65 It is thought that this exposure to the

mothers diet continues via breast feeding.66 However it is not known how far this applies to babies who are tube fed who dont have frequent close contact with their mothers. In summary; it is not known how any sensitive period for taste is affected by preterm birth. However, from data on term infants it seems highly likely that the later a preterm infant is introduced to new tastes the less likely they are to accept a wide variety of foods. A continued emphasis on offering a wide variety of foods in an appropriate social context is likely to be of long-term benet.67 Acceptance of new textures From around 6 months to early childhood chewing is continuously developed.47,68 Observational evidence suggests that this skill is aided by provision of foods of an appropriate texture.33 There is discussion around the introduction of solids to term infants with a move towards allowing the infant to take and eat food itself when he/she is ready, often termed baby led weaning.69 This has developed for many reasons one being that there does not seem to be any evidence to support puree as the rst food texture in healthy term infants. With this approach rst foods can be quite highly textured. However preterm infants will often not have the skills to begin weaning in this way. To ensure safety with baby led weaning it is recommended that a term infant is able to sit unaided, grasp food and take it to their mouth, actions which may be delayed in a preterm infant. The proponents of baby led weaning speculate that this approach encourages infants to take a wider variety of foods. Future research in this area with respect to term, and an appropriately modied approach preterm infants, would be of interest as there may be advantages for both groups. The later that weaning is started the more rapidly an infant should progress through textures with the promotion of nger foods as soon as the infant shows cues that they are ready to try them. This may help develop hand eye co-ordination and allow the child more control during feeding. As with term infants it is essential that the infant is never left unattended when eating. It has been suggested that delay in introduction of lumpy foods can lead to feeding problems if a sensitive period is missed, probably between 6 and 9 months.70e73 These were all observational studies and therefore do not conclusively prove this relationship, nevertheless the well accepted concept of programming during a sensitive period in development would explain it.33 The introduction of lumpy food should not be delayed until the appearance of teeth as they are not needed by the infant for chewing.32 The ability to chew foods is probably aided by giving foods of the appropriate texture at the appropriate time. Although the theory of a critical or sensitive time for acceptance of lumps has not been tested in a randomised controlled trial there are many anecdotal observations of feeding problems following delay of the usual weaning pattern - due to severe chronic illness, surgery, nasogastric feeding or gastrostomy feeding. In summary; it seems prudent to advise that most preterm infants should at least be offered lumps before 9 months uncorrected age. Some infants may progress rapidly with increasingly textured food, others with more delayed development will need to go more slowly e individualised advice is essential. Sleeping time More frequent waking for feeds and particularly night waking is often used as a cue for weaning. However there is evidence from

PAEDIATRICS AND CHILD HEALTH 19:9

407

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

studies in term infants that starting solids did not alter sleep patterns. One study randomised babies to solids at bed time from either 4 weeks or 4 months after birth and found no difference in sleep time between groups.74 Another was observational and found that age of giving solids was not related to the frequency of night feeding.24 Although a review of the literature found that breast fed term infants were more likely to be sleeping through the night at 9 months if they had been weaned before 12 weeks, this effect wasnt seen in formula fed infants, and in preterm infants there was no relationship between time of weaning and sleep in either breast or formula groups.75 Conclusion Behavioural and developmental issues may be more important to consider than nutritional ones when deciding why to wean preterm infants; although close attention is also needed to ensure nutritional adequacy through the weaning period. It is suggested that most preterm infants may be ready to wean some time between 5 and 8 months from birth.

Potential problems perceived with weaning between 5 and 8 months from birth
Weaning between 5 and 8 months from birth may appear to some to be potentially problematic as it may be considered too early. The following is an examination of problems which have been suggested and exploration of the strength of the evidence for them, and a nal summary. Gastrointestinal Immature gut barrier function In preterm infants gut barrier function has been assessed both by milk protein and sugar absorption; with the former showing increased absorption by the preterm gut intact into the circulation76 and the latter that the gut barrier appears to adapt rapidly post natally independent of gestational age or birth weight.77 There is work suggesting that in infants born at 32e36 weeks gestation, gut permeability is reduced by the 30th postnatal day.78 Earlier it was shown that by 37 post-conceptual weeks, macromolecular absorption in premature infants was similar to those born at term during their rst postnatal month.79 Animal studies suggest that the current practice of giving antenatal steroids to a mother threatening preterm delivery may accelerate intestinal mucosa, as well as lung maturation,80 although this has not been shown in humans. The initiation of enteral feeding may enhance this process as it is associated with an earlier decrease in intestinal permeability81 and improved mineral absorption.82 This evidence indicates that the maturation of the preterm infants gut barrier function occurs well before weaning would be considered and is accelerated by enteral feeds and possibly by the administration of ante natal steroids. Immature digestive capacity Despite their immaturity the majority of preterm infants are digesting and absorbing whole proteins, lactose and dietary fats well before they reach term due date; some many months earlier. Preterm infants are able to digest simple carbohydrate and protein from early on postnatally.83e86 Pancreatic function in infants born <28 weeks gestation may take 2 weeks to reach normal levels.87 Baseline pepsinogen is lower in preterm

compared to term infants however there is evidence that it is secreted appropriately in response to enteral nutrition.88 Pepsin activity is dependent on a pH below 5, in moderately premature infants gastric acid secretion appears sufcient.89,90 External factors such as medication which raises stomach pH above 5 may adversely affect protein digestion.91 Fat digestion will be the most compromised early on as pancreatic lipase takes the longest to mature. However, there is evidence that ability to digest fat is adequate in preterm infants.92,93 Rate of maturation is dependent on post conceptual age, postnatal age and timing of enteral feeding.94 Dietary fat can enhance gastric lipase in preterm infants95: gastric and lingual lipase being especially important in the preterm infant.96 As breast milk contains both amylase and bile salt stimulated lipase,97,98 breast-fed infants have additional mechanisms to digest starch and lipid. The human fetus demonstrates low lactase levels up until late in gestation. However, in infants born as early as 26 weeks post conception, enteral feeding, particularly with human milk leads to a signicant rise in lactase activity.99,100 In summary the digestive capacity of the preterm infant starts lower than in the term born counterpart but maturation accelerates post natally to give adequate function by the time weaning is likely to occur. Immature motor function The post-prandial motor response of the neonates small intestine has been shown to be related to volume of enteral bolus and number of days enteral feeds have been given: post conceptional age was not an important determinant of motor response.101 Similarly early enteral feeding accelerates the development of mature gut motility patterns compared to late fed control preterm infants.102 This indicates that development of gut motor function is accelerated post delivery, whatever the gestation of the infant, as long as enteral feeds are given. Insufcient gut hormones In fetal mammals amniotic uid is swallowed and exerts a trophic effect on the stomach ensuring normal gastric acid secretion.103 Therefore being kept nil by mouth for long periods post delivery is unphysiological and risks delaying post natal gut maturation. If enteral feeding is instituted shortly after birth even minimal feedings in preterm infants can lead to a signicant increase in gut hormones.104 This does not occur in infants given intravenous nutrition alone.105 Summary Although the human gut may develop according to a preset biological clock, up until approximately 25 weeks gestation106,107 all the evidence points to precocious maturation as a response to the trigger of enteral feeding; particularly if human milk is used rather than formula. A preterm enterally fed infant will have a more mature gut once it reaches term date than its counterpart born at term. Renal immaturity The number of nephrons continues increasing in preterm infants until approximately 36 weeks post-conceptual age. However individual functional maturation occurs from birth, be it term or preterm, triggered by the contraction of extra cellular water which occurs post natally.108 This precocious renal maturation

PAEDIATRICS AND CHILD HEALTH 19:9

408

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

suggests adequate preparation for weaning in preterm infants. One study has indicated normal serum biochemistry in preterm infants weaned from 2 weeks corrected age.5 This is well before it is recommended that weaning should occur, and as with term infants it is important to ensure that carers do not give food with an unacceptably high renal solute load.

It seems likely that preterm infants are no more likely to develop atopic disease than term infants. The current guidelines for babies at low risk of allergy is to introduce any foods from the start of weaning, as long as they are low in salt.128 However, if there is a family history of atopy a more cautious approach is recommended.129 It seems reasonable to apply the same strategies to preterm infants. Developmental and oro motor immaturity Head control and trunk stability provide the necessary gross motor skills for the ne motor control of hands and mouth needed during weaning. There is evidence that motor development matures according to post conceptional rather than post natal age and that both ne and gross motor development are adversely affected by preterm birth.130 Preterm infants who have prolonged illnesses often have poor somatic growth, which can contribute further to gross motor delay; resulting in poor head control, reduced trunk stability and immature oral-motor skills. It is useful to wait until a certain level of motor and developmental maturity has been achieved prior to weaning: head control being important, however many preterm infants will not be able to sit unaided even though other cues are present indicating that weaning could start. There is some evidence that waiting until around 3e4 months corrected age will allow some progression with motor skills ensuring adequate head control; as all preterm infants appear to have delayed motor development130; however individual assessment of cues remains essential. In high risk preterm infants the majority have been observed to hold their head in the midline at 4 months corrected and well over half able to sit supported at this age. There have been observations that some carers may delay weaning foods of a more challenging texture as they assume their infant is not able to manage them, presumably due to motor delay.46 However these infants may in fact be able and ready to move onto more textured food, and would benet from the opportunity to start experiencing these foods. Mothers of more robust term infants give them more opportunity for motor activity e so may also allow them more self feeding opportunities131 this may also apply to mothers of preterm infants. Concerns have been raised that weaning an infant with immature oro motor function constitutes passive over, or force feeding.6,132 However term infants as young as 4 months and certainly by 6 months can direct feeding interactions by showing food preferences and satiety,40,45 and there is some evidence that preterm infants can also do this.46 Feeding skills are continuously rened over the early years as infants move towards a more adult type diet. The risk of leaving weaning too late must be balanced against the risks of passive over feeding. If weaning is left too late progress may be more difcult due to the overshooting of periods where the infant is receptive to change. To reduce the risk of feeding problems it is important that the person feeding prepares an optimal environment.133 Caregiver/ infant interaction is vital, the caregiver has the responsibility to closely observe and react to the infants cues to recognise both hunger and satiety.

High risk of allergy It has often been suggested that preterm infants are at increased risk of food allergy due to immaturity of the gut and immune system, however, if anything evidence points to the opposite. A large epidemiological study found that in a group of 443 children with atopic eczema there was a signicant lack of subjects born before 37 weeks gestation.109 Others have either conrmed that preterm infants have a lower incidence of eczema compared to term infants110e112 or found no difference.113e116 One group looked at low birth weight and prematurity separately and found only fetal growth restriction to be associated with increased risk of atopic disease.117 Preterm infants have been found to have low levels of antibodies to cows milk118,119 and gliadin.118 In a randomised controlled study of weaning in preterm infants unpublished data on skin prick tests at 18 months found no difference in results between those weaned at a mean age of 15 vs. 18 weeks uncorrected; around half of the original cohort of 68 were tested with only 2 positives (Foote,K & Marriot, L personal communication). A prospective randomised study of hydrolysed protein preterm formula in infants at high risk of allergy has been reported. Randomisation to hydrolysed formula in the neonatal period failed to show any reduction in the incidence of allergic disease at either 4 or 12 months, although the risk of atopic dermatitis was reduced at 12 months.120 There are some studies which have suggested that preterm infants are at increased risk of allergies. In a group of small for gestational age infants, of which an unspecied number were preterm a high level of allergy was reported at 12 months.121 Another group has shown signicantly more allergic symptoms cumulatively to the age of 11 years in human milk fed preterm infants compared to those fed formula, however, incidence was not different between groups or the general population at 11 years.122 While another found that preterm infants with a strong family history of atopy were more at risk of atopic symptoms; there was no link with breast feeding or the time of starting allergenic foods.123 In a group of preterm infants followed prospectively introduction of 4 or more foods before 17 weeks post term was associated with higher risk of eczema at one year.124 However the authors and an accompanying editorial pointed out that at that time an authoritative statement on the introduction of solids and the development of allergy was still not possible,125 and certainly not for preterm infants. Preliminary data from a further study showed a higher incidence of atopy at 8 months in those weaned early (at 12 rather than 16 weeks), but at 12 months there was no difference.7 Preterm infants do seem to be at more risk of wheeze126 (in this study atopy was only associated with later not early wheeze) this may be due to lung damage in the neonatal period.127

PAEDIATRICS AND CHILD HEALTH 19:9

409

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

Obesity Some have suggested that early introduction of solids to preterm infants, particularly cereals, could result in obesity.6 Although this has been a concern in term infants it has not been borne out by longitudinal follow-up studies.75,134 Obesity is a problem rarely seen in preterm infants whatever their age at weaning. In fact, despite reporting early introduction of cereals in a preterm cohort no problem with obesity was reported.6 The time when rapid weight gain is often seen is the period soon after discharge, before weaning has started. Compromised nutritional density of diet This is a risk as many of the traditional rst weaning foods have a lower energy density than the milk they will replace. Italian mothers of preterm infants were found to start with fruit and didnt move onto meat until 5 weeks after the initiation of weaning,11 this would reduce the nutrient density of the diet as the infants would have been likely to reduce the volume of milk taken once solids were started.5 In young malnourished children it has been suggested that an minimum energy density of 0.7 kcal/g (roughly equivalent to breast milk) is needed to optimise catch up growth.135 Meals must also be offered sufciently frequently to allow enough food to be eaten.135 Intervention, in the form of a randomised controlled trial to promote early weaning along with specic advice to optimise nutrient density of the diet, lead to greater energy intake and greater weight and length at 6 months compared to a control group.20 It appears then that introduction of solids leads to a risk of reduction in dietary nutrient density, which could compromise growth. Advice to ensure maintenance of nutrient density in the weaning diet is advisable. In those infants where poor growth is an issue dietetic supervision during weaning is recommended. Summary In the majority of cases there is not sufcient evidence to back the suggestion that weaning between 5 and 8 months from birth will put the preterm infant at risk.

If the infant has an immature suckle pattern in response to solids, smooth puree is easy to suck and swallow and should reduce risk of gagging or vomiting. A gradual progression through to family foods is then recommended.143 Foods prepared at home are easier to modify with respect to texture as well as acclimatising the infant to the avours of the families usual foods. Healthy preterm infants will in general progress normally to more textured foods and may not need to have puree food for long. In those who have undergone a particularly stormy course and who establishment of oral feeding has been delayed, an early programme of oral desensitisation to promote ingestion of both milk and solids may be necessary. Gastro-oesophageal reux disease Preterm infants are at higher risk of gastro-oesophageal reux disease (GORD) and this itself has been linked with feeding problems.139,144,145 Although interestingly one observation found that GORD was not associated with sucking problems in preterm infants.146 There is evidence that nausea,147 pain148 and gastrointestinal illness149,150 are the prime determinants of feeding problems rather than the vomiting they are often associated with. Therefore with respect to GORD acid reux is the prime factor to control to avoid the pain of oesophagitis, whereas non acid reux may not be a cause for concern in the majority of infants. Term infants with GORD have been found to have signicantly more oral motor dysfunction and less age appropriate feeding skills compared to controls.145 This suggests that in preterm infants with conrmed GORD a gradual progression through purees and onto lumps would be of benet. As such infants may take longer to reach feeding milestones they are a group who may benet from weaning at the earlier end of the suggested age range. However this has not been formally evaluated and would benet from further study. There has been a suggestion that starting solids early may be useful to reduce GORD symptoms, thought there is little evidence to back this view. It may have developed through the observation that spontaneous resolution of symptoms often occurs around the time of weaning. In fact there is a risk that solids may exacerbate GORD by prolonging gastric emptying. This risk needs to be balanced with the possible advantage of weaning at an earlier age as discussed above.

Special considerations for sick preterm infants

Feeding problems Prevalence Preterm infants are at high risk of feeding problems.45,136e140 Some evidence suggests that this is linked to severity of illness in the perinatal period rather than prematurity per se,141,142 although not all.12 An assessment at 6 months corrected age of babies born preterm found a range of signicant ongoing feeding difculties and aversive behaviours compared to babies born at term. Counter intuitively their mothers didnt report any signicant feeding aversion or negative feelings around feeding. However they were more likely to be using pureed food, and therefore have more rapid, less stressful meal times than if more challenging textures were offered.45 Delayed coordination may lead to a longer time for feeding development and self feeding: in these circumstances smooth puree is the easiest texture to offer initially.

Respiratory compromise In preterm infants respiratory compromise has been associated with high risk of feeding problems.137e139 Early endotracheal intubation can lead to poorer oral feeding many weeks after extubation.151 Infants who develop chronic lung disease (CLD) can present particular problems. As they must stop breathing while swallowing and as respiration takes precedence over feeding, infants with respiratory problems are often poor feeders.152 Results of a questionnaire highlight the feeding problems to which infants with CLD are prone when on solids.137 In this study infants with CLD were weaned later than controls; the author speculates that this delay may have put them beyond a sensitive period for acceptance of solids. These infants often need multi disciplinary input to help progress with oral nutrition, it may be particularly important to

PAEDIATRICS AND CHILD HEALTH 19:9

410

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

ensure the timely introduction of solids to enhance acceptance later on. Prior nasogastric tube feeding In infants, term or preterm, prolonged early nasogastric tube feeding is associated with feeding difculties.153e155 Placement of a nasogastric feeding tube for >3 weeks has been linked to delayed feeding skills and altered oral sensitivity in healthy preterm infants at 11e17 months.156 Chronic low grade inammation around the tube placement site has been suggested as a possible aetiological factor.71 Infants who had at least 2 weeks of non oral nutrition during their rst 3 months were found to have abnormal physiological responses to touch (assessed at 3e18 months) including gagging and distress.157 Nutritional factors Preterm infants with ongoing chronic problems will be more at risk of a range of nutritional problems and would benet from regular assessment to identify where they are particularly at risk. The highest risk is poor energy and protein intake however assessment of intake of certain micronutrients is also advisable. Zinc stores in preterm infants could be low depending on nutritional management and severity of illnesses in early neonatal period,158 low stores could limit growth therefore a balanced weaning diet is essential. Iron stores are low in all preterm infants and it is important to ensure that they receive good dietary sources and not rely on supplements. Calcium and phosphorus input seems to inuence how rapidly bone mineralisation is normalised following preterm birth159e161; although some spontaneous resolution appears to occur.162 Those on lower mineral intake still do catch-up with term infants, but with up to a 2 year lag,163 even exceeding the bone mineral content of their term counterparts by 5 years.164 In addition it has been found that once into adulthood early mineral input did not positively affect bone mineral status,165 nevertheless it seems prudent to normalise bone mineral content as early as possible. Although calcium and phosphorus are normally adequate from dairy sources, breast fed infants require a vitamin D supplement. If a supplement of vitamin D is needed it is usually given as part of a multivitamin as there is currently no suitable single vitamin D supplement. Caregiver factors The severity of perinatal risks partly predicts an infants sleeping and eating problems. However the intensity of parents post traumatic reactions was an important independent factor.166 In a study of preterm infants born in the 1990s early weaning was associated with less positive health behaviours in the mother.8 Early food refusal has been linked to psychosocial problems in the family and less positive perceptions of parenting by the parents167; however it is often not possible to say which came rst e the infants feeding problem or the parents negative perceptions. When a child is not given sufcient control over own intake this can disrupt the childs acceptance of a variety of foods and decrease awareness of satiety cues.168 Caregivers should be given the support necessary to enable them to be responsive to their childs cues and needs.

Positioning Caregivers have been reported to have difculties in positioning preterm babies for feeding.45 This may be as a result of inability of the infant to stabilise the head, neck, shoulders and trunk through immaturity and through delay following prolonged illness. Guidance on appropriate positioning should help facilitate weaning and ensure that weaning is not unnecessarily delayed. Infants need the opportunity to touch and feel food to start to develop the skills to feed themselves. Where ever possible nger foods should be encouraged as they are often better tolerated than food from a spoon, possibly because with nger foods children are more in control of what goes in their mouths. Finger foods should start when the infant begins to be able to hold objects and starts to bring them to their mouths. For infants who nd bringing their hands to the mid line difcult it may be useful to have physiotherapy and/or occupational therapy involvement.

Weaning guidelines
Previous guidelines Government guidelines on weaning from 1994 include a section on preterm infants with several recommendations,1 the advice is summarised as follows: 1. Wean once at least 5 kg achieved 2. No tongue thrust reex 3. Able to eat from spoon 4. Weaning diet must continue to ensure nutritional adequacy in particular with respect to iron, zinc, calcium, phosphorus and energy. The relevance of the above four points are considered in the light of evidence previously discussed. 1. If weaning is left until 5 kg some of those born earliest who are growing below the 2nd centile could be 10 months uncorrected age before they reach this weight and risk being beyond or towards the end of sensitive periods for texture and taste. Therefore it is suggested that setting a weight criterion is not a helpful guide. 2. Loss of tongue thrust (protrusion) reex is not a useful indicator of when a preterm infant is ready for weaning, and often persists during the initial stages of spoon feeding. It is believed that reduction of this reex may, in fact, be aided by introduction of solids. 3. Ability to eat from a spoon involves a degree of lip seal. However this does not develop until after weaning in term infants. Preterm infants may have delayed development of skills preventing lip seal despite being ready to wean in other respects. If weaning is delayed until a mature pattern of spoon-feeding is possible, other sensitive periods for acceptance of solids may be missed. The development of lip seal may be aided by provision of spoon feeds. 4. The need to ensure nutritional adequacy of the weaning diet is fully endorsed. Suggested new guidelines Previous guidelines may no longer be appropriate so new ones are needed to accommodate the specic needs of the current preterm population. Preterm infants, particularly those who have had a complex neonatal course are at high risk for feeding problems and delayed development of appropriate feeding skills.

PAEDIATRICS AND CHILD HEALTH 19:9

411

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

Many of these problems can be addressed by starting weaning at a time when infants are showing readiness cues. Correcting for prematurity and following the recommendations given for healthy term infants will often result in delaying weaning beyond 10 months uncorrected age which may compromise nutrition and delay key opportunities for development of feeding skills. A graded introduction of textures will allow preterm infants at high risk of feeding problems to develop feeding skills at their own pace. Well preterm infants may be able to proceed with textures more rapidly. It is likely that the majority of infants born prematurely will be ready to start weaning somewhere between the ages of 5e8 months from the date of birth (uncorrected). However there is evidence that the motor development necessary for safe and successful transition to solid foods may not have been achieved until at least 3 months corrected age. The exact age will vary from one infant to another so advice given on the cues and physical development required is needed. Such advice is available in a separate document available on the dietitians web page at www.bapm.org. This is a joint consensus statement which has been composed by a group of UK paediatric dietitians and speech and language therapists. It is based on a review of the literature, Delphi questionnaire results and consensus meeting held in London on 26th June 2007. There is also practical advice in the form of a booklet produced by Bliss, a UK based charity for sick and preterm infants www.bliss.org.uk. Although prematurity is usually dened as <37 weeks those well preterm infants born closer to term (i.e.>34 weeks) may be ready to start weaning at approximately 6 months of age (corrected) as advised for infants born at term.

Conict of interest
No conict of interest declared.

REFERENCES 1 COMA. Weaning and the weaning diet. London: Department of Health, HMSO; 1994. 2 Department of Health. Infant feeding recommendation. Available at: http://www.dh.gov.uk/assetRoot/04/09/69/99/04096999.pdf; 2004. 3 Rea J. J Neonatal Nurs 1997; 3: 19e23. 4 Fanaro S, Vigi V. J Pediatr Gastroenterol Nutr 2007; 45(Suppl. 3): S204e9. 5 DSouza SW, Vale J, Sims DG, et al. Arch Dis Child 1985; 60(3): 215e8. 6 Ernst JA, Bull MJ, Rickard KA, et al. J Pediatr 1990; 117(2 Pt 2): S156e66. 7 Hampton SM. Proc Nutr Soc 1999 Feb; 58(1): 75e8. 8 Fewtrell MS, Lucas A, Morgan JB. Arch Dis Child Fetal Neonatal Edn 2003; 88(4): F296e301. 9 Norris FJ, Larkin MS, Williams CM, et al. Eur J Clin Nutr 2002; 56(5): 448e54. 10 Cooke RJ, Grifn IJ, McCormick K, et al. Pediatr Res 1998 Mar; 43(3): 355e60. 11 Fanaro S, Borsari G, Vigi V. J Pediatr Gastroenterol Nutr 2007; 45(Suppl. 3): S210e4. 12 Buswell CA, Leslie P, Embleton ND, et al. Dysphagia 2009; 24(1): 20e5. 13 SACN. Minutes of the Scientic Advisory Group on Nutrition (SACN) meeting on Jan 27th 2001. Available at: www.sacn.gov.uk; 2001.

14 Gartner LM, Morton J, Lawrence RA, et al. American Academy of Pediatrics. Pediatr 2005; 115(2): 496e506. 15 Aggett PJ, Barclay S, Whitley JE. Acta Paediatr Scand Suppl 1989; 361: 1965e91. 16 Pisacane A, De Vizia B, Valiante A, et al. J Pediatr 1995; 127(3): 429e31. 17 Tsang RC, Uauy R, Koletzko B, et al. 45243. 2nd edn. Cincinnati Ohio: Digital Educational Publishing Inc; 2005. 18 Grifn IJ, Cooke RJ, Reid MM, et al. Arch Dis Child Fetal Neonatal Edn 1999; 81(1): F45e9. 19 Domellof M, Lonnerdal B, Abrams SA, et al. Am J Clin Nutr 2002; 76(1): 198e204. 20 Marriott LD, Foote KD, Bishop JA, et al. Arch Dis Child Fetal Neonatal Edn 2003; 88(4): F302e7. 21 Reilly JJ, Wells JC. Br J Nutr 2005; 94(6): 869e72. 22 Islam MM, Peerson JM, Ahmed T, et al. Am J Clin Nutr 2006; 83(4): 851e8. 23 Capdevila F, Vizmanos B, Marti-Henne C. J Am Coll Nutr 1998; 17(3): 256e62. 24 Heinig M, Nommsen LA, Peerson JM, et al. Acta Paediatr 1993; 82: 999e1006. 25 Cohen RJ, Brown KH, Canahuati J, et al. Lancet 1994; 344(8918): 288e93. 26 Mehta KC, Specker BL, Bartholmey S, et al. Pediatrics 1998 Sep; 102(3 Pt 1): 569e73. 27 Carruth BR, Skinner JD, Houck KS, et al. J Am Coll Nutr 2000; 19(3): 405e12. 28 Simondon KB, Gartner A, Berger J, et al. Am J Clin Nutr 1996 Oct; 64(4): 537e45. 29 Michaelsen K, Jorgensen MH. Eur J Clin Nutr 1995; 49: 467e83. 30 Dewey KG, Cohen RJ, Brown KH, et al. Am J Clin Nutr 1999; 69(4): 679e86. 31 Arvedson J, Rogers B, Brodsky L. Anatomy, embryology and physiology. In: Arvedson J, Brodsky L, eds. Pediatric swallowing and feeding assessment and management; 1993. p. 46. 32 Bosma. Clin Nutr 1986; 5: 210. 33 Stevenson RD, Allaire JH. Pediatr Clin North Am 1991; 38(6): 1439e53. 34 Morris Evans. Pre-feeding skills. Tuscon Arizona: Therapy Skills Builders; 1987. 35 Mennella JA, Beauchamp GK. Nutr Rev 1998 Jul; 56(7): 205e11. 36 Steiner JE. Adv Child Dev Behav 1979; 13: 257e95. 37 Kajiura H, Cowart BJ, Beauchamp GK. Dev Psychobiol 1992; 25(5): 375e86. 38 Mennella JA, Grifn CE, Beauchamp GK. Pediatrics 2004; 113(4): 840e5. 39 Beauchamp GK, Cowart BJ, Moran M. Dev Psychobiol 1986; 19(1): 17e25. 40 Harris G, Thomas A, Booth DA. Dev Psychobiol 1990; 26: 534e8. 41 Kare MR, Beauchamp GK. Am J Clin Nutr 1985; 41(2 Suppl): 418e22. 42 Fomon SJ, Thomas LN, Filer Jr LJ. J Pediatr 1970; 76(2): 242e6. 43 Sullivan SA, Birch LL. Pediatrics 1994; 93(2): 271e7. 44 Hill DL, Mistretta CM. Trends Neurosci 1990; 13(5): 188e95. 45 Mathisen B, Worrall L, OCallaghan M, et al. Adv Speech Lang Pathol 2000; 2(1): 9e17. 46 Pridham K, Steward D, Thoyre S, et al. Early Hum Dev 2007; 83(5): 293e305. 47 Gisel EG. Dev Med Child Neurol 1991; 33(1): 69e79. 48 Reilly S, Skuse D, Mathisen B, et al. Dysphagia 1995; 10(3): 177e91.

PAEDIATRICS AND CHILD HEALTH 19:9

412

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

49 Gerrish CJ, Mennella JA. Am J Clin Nutr 2001; 73(6): 1080e5. 50 Maier AS, Chabanet C, Schaal B, et al. Clin Nutr 2008; 27(6): 849e57. 51 Birch LL, McPhee L, Shoba BC, et al. Appetite 1987; 9(3): 171e8. 52 Guthrie CA, Rapoport L, Wardle J. Appetite 2000; 35(1): 73e7. 53 Birch LL. Annales Nestle 1998; 56(1): 11e8. 54 Forestell CA, Mennella JA. Pediatrics 2007; 120(6): 1247e54. 55 Harper L, Sanders K. J Exp Child Psychol 1975; 20: 206e14. 56 Birch LL, Zimmerman S, Hind H. Child Dev 1980; 51: 856e61. 57 Rozin P. Dev Psychol 1990; 26(4): 555e62. 58 Pliner P, Pelchat ML. Appetite 1986; 7(4): 333e42. 59 El-Sohemy A, Stewart L, Khataan N, et al. Forum Nutr 2007; 60: 176e82. 60 Keskitalo K, Knaapila A, Kallela M, et al. Am J Clin Nutr 2007; 86(1): 55e63. 61 Rajan L. Health Visitor 1986; 59(2): 41e4. 62 Wardle J, Cooke L. Br J Nutr 2008; 99(Suppl. 1): S15e21. 63 Birch LL. Dev Psychol 1990; 26: 515e9. 64 Beauchamp G, Mennella J. Annales Nestle 1998; 56(1): 19e31. 65 Mennella JA, Johnson A, Beauchamp GK. Chem Senses 1995; 20(2): 207e9. 66 Mennella JA. J Hum Lact 1995 Mar; 11(1): 39e45. 67 Cowart BJ. Psychol Bull 1981; 90(1): 43e73. 68 Green J, Moore C, Ruark J, et al. J Neurophysiol 1997; 77: 2704e16. 69 Rapley G, Murkett T. 1st edn. Vermillion; 2008. ISBN: 978-0-09192380-8. 70 Illingworth R, Lister J. J Pediatr 1964; 65: 839e49. 71 Skuse D. Arch Dis Child 1993; 69(5): 604e8. 72 Northstone K, Emmett P, Nethersole F. J Hum Nutr Diet 2001 Feb; 14(1): 43e54. 73 Coulthard H, Harris G, Emmett P. Matern Child Nutr 2009; 5(1): 75e85. 74 Macknin ML, Medendorp SV, Maier MC. Am J Dis Child 1989; 143(9): 1066e8. 75 Morgan JB, Lucas A, Fewtrell MS. Arch Dis Child 2004; 89(8): 728e33. 76 Roberton DM, Paganelli R, Dinwiddie R, et al. Arch Dis Child 1982; 57(5): 369e72. 77 van Elburg RM, Fetter WP, Bunkers CM, et al. Arch Dis Child Fetal Neonatal Edn 2003; 88(1): F52e5. 78 Martinez-Augustin O, Boza- JJ, Del-Pino J, et al. Biol Neonate 1997; 71(4): 215e23. 79 Axelsson I, Jakobsson I, Lindberg T, et al. Acta Paediatr Scand 1989; 78: 532e7. 80 Pang KY, Newman AP, Udall JN, et al. Am J Physiol 1985; 249(1 Pt 1): G85e91. 81 Shulman RJ, Schanler RJ, Lau C, et al. Pediatr Res 1998 Oct; 44(4): 519e23. 82 Schanler RJ, Schulman RJ, Lau C. Pediatr Res 1995; 37: 319A. 83 Senterre J. Acta Paediatr Scand 1980; 69(5): 653e7. 84 Lebenthal E, Lee PC. Pediatrics 1980; 66(4): 556e60. 85 Kolacek S, Puntis JW, Lloyd DR, et al. Arch Dis Child 1990; 65(2): 178e81. 86 Nissler K, Von Katte I, Huebner A, et al. J Pediatr Gastroenterol Nutr 2001; 33(1): 28e31. 87 Kori M, Maayan-Metzger A, Shamir R, et al. Arch Dis Child Fetal Neonatal Ed 2003 Mar; 88(2): F106e8. 88 Yahav J, Carrion V, Lee PC, et al. J Pediatr 1987; 110(6): 949e51. 89 Hyman PE, Clarke DD, Everett SL, et al. J Pediatr 1985 Mar; 106(3): 467e71.

90 Kelly EJ, Newell SJ, Brownlee KG, et al. Early Hum Dev 1993 Dec 31; 35(3): 215e20. 91 Hamosh M. Digestion in the newborn. Clin Perinatol 1996 Jun; 23(2): 191e209. 92 Rings E, Minich D, Fetter W, et al. Pediatr Res 1999; 45: 290A. 93 Deenamamode J, Nicholl R, Sherwood R, et al. Pediatr Res 1998; 44: 435A. 94 Forget P, Van den Neucker A, Degraeuwe P, et al. Pediatr Res 1995; 37: 307A. 95 Pereira G, Barsky D, Hamosh M, et al. Pediatr Res 1995; 37: 316A. 96 Hamosh M. A review. Pediatr Res 1979; 13(5 Pt 1): 615e22. 97 Lindberg T, Skude G. Pediatrics 1982; 70(2): 235e8. 98 Heitlinger LA, Lee PC, Dillon WP, et al. Pediatr Res 1983; 17(1): 15e8. 99 Shulman RJ, Schanler RJ, Lau C. Pediatr Res 1996; 39: 320A. 100 Shulman RJ, Wong WW, Smith EO. Am J Clin Nutr 2005 Feb; 81(2): 472e9. 101 Bissett WM, Watt J, Rivers RPA, et al. Arch Dis Child 1989; 64: 1356e61. 102 Berseth CL. J Pediatr 1992 Jun; 120(6): 947e53. 103 Mulvihill SJ, Stone MM, Fonkalsrud EW, et al. J Surg Res 1986; 40(4): 291e6. 104 Lucas A, Bloom SR, Aynsley Green A. Acta Paediatr Scand 1986; 75(5): 719e23. 105 Lucas A, Bloom SR, Aynsley Green A. Acta Paediatr Scand 1983; 72(2): 245e9. 106 Newell S. Semin Neonatol 1996; 1: 59e66. 107 Newell SJ. Clin Perinatol 2000 Mar; 27(1): 221e34. viii. 108 Aperia A, Zetterstrom R. Clin Perinatol 1982; 9(3): 523e33. 109 David TJ, Ewing CI. Arch Dis Child 1988; 63(4): 435e6. 110 Buhrer C, Grimmer I, Niggemann B, et al. Lancet 1999 May 15; 353(9165): 1674. 111 Siltanen M, Kajosaari M, Pohjavuori M, et al. J Allergy Clin Immunol 2001 Feb; 107(2): 229e34. 112 Agosti M, Vegni C, Gangi S, et al. Acta Paediatr Suppl 2003; 91(441): 44e7. 113 Klebanoff MA, Berendes HW. Arch Dis Child 1988; 63(12): 1519e20. 114 de Martino M, Donzelli GP, Galli L, et al. Biol Neonate 1989; 56(6): 301e5. 115 Lucas A, Brooke OG, Morley R, et al. BMJ 1990; 300(6728): 837e40. 116 Kvenshagen B, Jacobsen M, Halvorsen R. Arch Dis Child 2009; 94(3): 202e5. 117 Steffensen FH, Sorensen HT, Gillman MW, et al. Epidemiology 2000 Mar; 11(2): 185e8. 118 Helms I, Rieger C. Eur J Pediatr 1987; 146: 131. 119 Siltanen M, Kajosaari M, Savilahti EM, et al. J Allergy Clin 2002; 110(4): 658e63. 120 Szajewska H, Mrukowicz JZ, Stoinska B, et al. Acta Paediatr 2004; 93(9): 1159e65. 121 Norris F, Williams P, Larkin M, et al. Ped Asthma Allergy Immunol 1999; 13(4): 169e75. 122 Savilahti E, Tuomikoski JP, Jarvenpaa AL, et al. Acta Paediatr 1993 Apr; 82(4): 340e4. 123 Forster J, Dungs M, Wais U, et al. Klin Padiatr 1990; 202(3): 136e40. 124 Morgan J, Williams P, Norris F, et al. Arch Dis Child 2004; 89(4): 309e14. 125 Khakoo GA, Lack G. Arch Dis Child 2004; 89(4): 295.

PAEDIATRICS AND CHILD HEALTH 19:9

413

2009 Elsevier Ltd. All rights reserved.

SYMPOSIUM: NUTRITION

126 Siltanen M, Savilahti E, Pohjavuori M, et al. Pediatr Pulmonol 2004 Jan; 37(1): 43e9. 127 Grischkan J, Storfer-Isser A, Rosen CL, et al. J Pediatr 2004; 144(3): 321e6. 128 Paediatric Group of the British Dietetic Association. Weaning infants onto solid foods. Position paper. Available at: http://www bda uk com/resources/statements/PositionStatementWeaning pdf; 2009. 129 Food Allergy & Intolerance Specialist Group of the British Dietetic Association. Practical dietary prevention strategies for infants at risk of developing allergic diseases. Professional consensus statement. www bda uk com. 130 van Haastert I, de Vries LS, Helders PJ, et al. J Pediatr 2006; 149(5): 617e22. 131 Wachs T. Monogr Soc Res Child Dev 1993; 83(7): 100e10. 132 Fomon SJ, Filer Jr LJ, Anderson TA, et al. Pediatrics 1979; 63(1): 52e9. 133 Satter E. J Pediatr 1990; 117(2 Pt 2): S181e9. 134 Burdette HL, Whitaker RC, Hall WC, et al. Am J Clin Nutr 2006; 83(3): 550e8. 135 Brown KH, Sanchez-Grinan M, Perez F, et al. Am J Clin Nutr 1995 Jul; 62(1): 13e8. 136 Douglas JE, Bryon M. Arch Dis Child 1996; 75(4): 304e8. 137 Martin M, Shaw NJ. J Human Nutr 1997; 10: 271e5. 138 Hawdon JM, Beauregard N, Slattery J, et al. Dev Med Child Neurol 2000; 42(4): 235e9. 139 Rommel N, De Meyer AM, Feenstra L, et al. J Pediatr Gastroenterol Nutr 2003; 37(1): 75e84. 140 Schadler G, Suss-Burghart H, Toschke AM, et al. Eur J Pediatr 2007; 166(8): 803e8. 141 Bazyk S. Am J Occup Ther 1990; 44(12): 1070e8. 142 Burklow KA, McGrath AM, Valerius KS, et al. Nutr Clin Pract 2002; 17(6): 373e8. 143 King C, Marriott L, Foote KD. Weaning your premature baby. London: BLISS. Available at: www.bliss.org.uk; 2008. 5th. 144 Field D, Garland M, Williams K. J Paediatr Child Health 2003; 39(4): 299e304. 145 Mathisen B, Worrall L, Masel J, et al. J Paediatr Child Health 1999; 35(2): 163e9. 146 veerman-wauters G, Palmer M, Goeleven D, et al. Gastroenterol 1999; 108(4): A760. 147 Pelchat ML, Rozin P. Appetite 1982; 3(4): 341e51.

148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168

Hyman PE. J Pediatr 1994; 125(6 Pt 2): S103e9. Johnson R, Harris G. J Reprod Inf Psychol 2004; 22(3): 183e8. Garb JL, Stunkard AJ. Am J Psychiatry 1974; 131(11): 1204e7. Blaymore-Bier JA, Ferguson A, Cho C, et al. Am J Dis Child 1993; 147(8): 858e62. Singer L, Martin RJ, Hawkins SW, et al. Pediatr 1992; 90: 380e4. Blackman JA, Nelson CL. Clin Pediatr Phila 1985 Aug; 24(8): 434e8. Senez C, Guys JM, Mancini J, et al. Childs Nerv Syst 1996 Oct; 12(10): 590e4. Mason SJ, Harris G, Blissett J. Dysphagia 2005; 20(1): 46e61. Dodrill P, McMahon S, Ward E, et al. Early Hum Dev 2004; 76(1): 23e37. Scarborough DR, Boyce S, McCain G, et al. Dev Med Child Neurol 2006; 48(6): 460e4. Askari A, Long CL, Blakemore WS. J Parenter Enteral Nutr 1979; 3(3): 151e6. Venkataraman P, Luthar HK, Neylan M. Pediatr Res 1990; 27: 293A. Chan GM. J Pediatr 1993; 123: 439e43. Bishop NJ, King FJ, Lucas A. Arch Dis Child 1993; 68: 573e8. Congdon P, Horsman A, Ryan SW, et al. Arch Dis Child 1990; 65: 1038e42. Schanler R, Burns PA, Abrams SA, et al. Pediatr Res 1992; 31: 583e6. Bishop NJ, Dahlenburg SL, Fewtrell MS, et al. Acta Paediatr 1996; 85: 230e6. Fewtrell MS, Williams JE, Singhal A, et al. Bone 2009; uncorrected proof. Pierrehumbert B, Nicole A, Muller-Nix C, et al. Arch Dis Child Fetal Neonatal Edn 2003; 88(5): F400e4. Lindberg L, Bohlin G, Hagekull B. Infant Ment Health J 1994; 15(3): 262e77. Birch LL, Fisher JO. Pediatrics 1998; 101(3 Pt 2): 539e49.

Acknowledgements
Many thanks to Annie Aloysius, Chris Jarvis, Gillian weaver and Trak Davis for very helpful comments and corrections.

PAEDIATRICS AND CHILD HEALTH 19:9

414

2009 Elsevier Ltd. All rights reserved.